An imprecise label ranking method for heterogeneous data

Tathagata Basu; Sébastien Destercke; Benjamin Quost

doi:10.1007/978-3-031-15509-3_5

An imprecise label ranking method for heterogeneous data

Tathagata Basu, Sébastien Destercke, Benjamin Quost

Civil And Environmental Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

Abstract

Learning to rank is an important problem in many sectors ranging from social sciences to artificial intelligence. However, it remains a rather difficult task to perform. Therefore, in some cases, it is preferable to perform cautious inference. For this purpose, we look into the possibility of an imprecise probabilistic approach for the Plackett-Luce model, a popular probabilistic model for label ranking. We aim at extending current Bayesian inference techniques for the Plackett-Luce model to an imprecise probabilistic setting so that we can deal with heterogeneous data by means of cautious mixture modelling. To achieve this, we perform a robust Bayesian analysis over a set of imprecise Dirichlet priors, which allows us to perform cautious label ranking. Finally, we use a synthetic dataset to illustrate our imprecise estimation method.

Original language	English
Title of host publication	Building Bridges Between Soft and Statistical Methodologies for Data Science
Editors	Luis A. García-Escudero, Alfonso Gordaliza, Agustín Mayo, María Asunción Lubiano Gomez, Maria Angeles Gil, Przemyslaw Grzegorzewski, Olgierd Hryniewicz
Place of Publication	Cham, Switzerland
Publisher	Springer
Pages	32-39
Number of pages	8
ISBN (Electronic)	9783031155093
ISBN (Print)	9783031155086
DOIs	https://doi.org/10.1007/978-3-031-15509-3_5
Publication status	Published - 25 Aug 2022
Event	10th International Conference on Soft Methods in Probability and Statistics - Valladolid, Spain Duration: 14 Sept 2022 → 16 Sept 2022

Conference

Conference	10th International Conference on Soft Methods in Probability and Statistics
Abbreviated title	SMPS 2022
Country/Territory	Spain
City	Valladolid
Period	14/09/22 → 16/09/22

Keywords

label ranking method
imprecise probabilistic approach
Plackett-Luce model

Access to Document

10.1007/978-3-031-15509-3_5

Cite this

Basu, Tathagata ; Destercke, Sébastien ; Quost, Benjamin. / An imprecise label ranking method for heterogeneous data. Building Bridges Between Soft and Statistical Methodologies for Data Science. editor / Luis A. García-Escudero ; Alfonso Gordaliza ; Agustín Mayo ; María Asunción Lubiano Gomez ; Maria Angeles Gil ; Przemyslaw Grzegorzewski ; Olgierd Hryniewicz. Cham, Switzerland : Springer, 2022. pp. 32-39

@inproceedings{31ec0ac53a33469dacaeb31d832eae55,

title = "An imprecise label ranking method for heterogeneous data",

abstract = "Learning to rank is an important problem in many sectors ranging from social sciences to artificial intelligence. However, it remains a rather difficult task to perform. Therefore, in some cases, it is preferable to perform cautious inference. For this purpose, we look into the possibility of an imprecise probabilistic approach for the Plackett-Luce model, a popular probabilistic model for label ranking. We aim at extending current Bayesian inference techniques for the Plackett-Luce model to an imprecise probabilistic setting so that we can deal with heterogeneous data by means of cautious mixture modelling. To achieve this, we perform a robust Bayesian analysis over a set of imprecise Dirichlet priors, which allows us to perform cautious label ranking. Finally, we use a synthetic dataset to illustrate our imprecise estimation method.",

keywords = "label ranking method, imprecise probabilistic approach, Plackett-Luce model",

author = "Tathagata Basu and S{\'e}bastien Destercke and Benjamin Quost",

year = "2022",

month = aug,

day = "25",

doi = "10.1007/978-3-031-15509-3\_5",

language = "English",

isbn = "9783031155086",

pages = "32--39",

editor = "Garc{\'i}a-Escudero, \{Luis A.\} and Alfonso Gordaliza and Agust{\'i}n Mayo and Gomez, \{Mar{\'i}a Asunci{\'o}n Lubiano\} and Gil, \{Maria Angeles\} and Przemyslaw Grzegorzewski and Olgierd Hryniewicz",

booktitle = "Building Bridges Between Soft and Statistical Methodologies for Data Science",

publisher = "Springer",

address = "Switzerland",

note = "10th International Conference on Soft Methods in Probability and Statistics, SMPS 2022 ; Conference date: 14-09-2022 Through 16-09-2022",

}

Basu, T, Destercke, S & Quost, B 2022, An imprecise label ranking method for heterogeneous data. in LA García-Escudero, A Gordaliza, A Mayo, MAL Gomez, MA Gil, P Grzegorzewski & O Hryniewicz (eds), Building Bridges Between Soft and Statistical Methodologies for Data Science. Springer, Cham, Switzerland, pp. 32-39, 10th International Conference on Soft Methods in Probability and Statistics, Valladolid, Spain, 14/09/22. https://doi.org/10.1007/978-3-031-15509-3_5

An imprecise label ranking method for heterogeneous data. / Basu, Tathagata; Destercke, Sébastien; Quost, Benjamin.
Building Bridges Between Soft and Statistical Methodologies for Data Science. ed. / Luis A. García-Escudero; Alfonso Gordaliza; Agustín Mayo; María Asunción Lubiano Gomez; Maria Angeles Gil; Przemyslaw Grzegorzewski; Olgierd Hryniewicz. Cham, Switzerland: Springer, 2022. p. 32-39.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

TY - GEN

T1 - An imprecise label ranking method for heterogeneous data

AU - Basu, Tathagata

AU - Destercke, Sébastien

AU - Quost, Benjamin

PY - 2022/8/25

Y1 - 2022/8/25

N2 - Learning to rank is an important problem in many sectors ranging from social sciences to artificial intelligence. However, it remains a rather difficult task to perform. Therefore, in some cases, it is preferable to perform cautious inference. For this purpose, we look into the possibility of an imprecise probabilistic approach for the Plackett-Luce model, a popular probabilistic model for label ranking. We aim at extending current Bayesian inference techniques for the Plackett-Luce model to an imprecise probabilistic setting so that we can deal with heterogeneous data by means of cautious mixture modelling. To achieve this, we perform a robust Bayesian analysis over a set of imprecise Dirichlet priors, which allows us to perform cautious label ranking. Finally, we use a synthetic dataset to illustrate our imprecise estimation method.

AB - Learning to rank is an important problem in many sectors ranging from social sciences to artificial intelligence. However, it remains a rather difficult task to perform. Therefore, in some cases, it is preferable to perform cautious inference. For this purpose, we look into the possibility of an imprecise probabilistic approach for the Plackett-Luce model, a popular probabilistic model for label ranking. We aim at extending current Bayesian inference techniques for the Plackett-Luce model to an imprecise probabilistic setting so that we can deal with heterogeneous data by means of cautious mixture modelling. To achieve this, we perform a robust Bayesian analysis over a set of imprecise Dirichlet priors, which allows us to perform cautious label ranking. Finally, we use a synthetic dataset to illustrate our imprecise estimation method.

KW - label ranking method

KW - imprecise probabilistic approach

KW - Plackett-Luce model

U2 - 10.1007/978-3-031-15509-3_5

DO - 10.1007/978-3-031-15509-3_5

M3 - Conference contribution book

SN - 9783031155086

SP - 32

EP - 39

BT - Building Bridges Between Soft and Statistical Methodologies for Data Science

A2 - García-Escudero, Luis A.

A2 - Gordaliza, Alfonso

A2 - Mayo, Agustín

A2 - Gomez, María Asunción Lubiano

A2 - Gil, Maria Angeles

A2 - Grzegorzewski, Przemyslaw

A2 - Hryniewicz, Olgierd

PB - Springer

CY - Cham, Switzerland

T2 - 10th International Conference on Soft Methods in Probability and Statistics

Y2 - 14 September 2022 through 16 September 2022

ER -

An imprecise label ranking method for heterogeneous data

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this